Detection And Characteristic Extraction About Traveling Wave For Overhead Lines

Feeding line is an important part of the traction power supply system, its running state determines the security and efficiency of the traction power supply system directly. But the power net especially the malfunction of the high voltage feeding line is inevitable. Once the feeding line goes wrong, it will bring a great loss to the traction line of power supply, the power equipment and the integral railway transmission. It's the date of the electrization railway speed up, the trains run fast and frequently, so if the malfunction hasn't been solved timely, the security of the railway running will be effected greatly, and the detection of the fault signal and feature extraction, distinguishing the fault type, reducing the time for malfunction, and making sure the railway secure are all have an important sense. At the same time ,with the railway lines increasing, its functions in cities' transport system are prominent and with the development of the science technology ,social civilization progressing , People have brought forward higher requests for security and comfort of urban railway transmission ,especially Beijing as a center of the national politics and culture, making sure the railway can run safely not only has some economical sense but also has very important political sense. People focus their eyes on the point of urban railway security and stabilization. As a basal part, the fault detection of the feeding line and feature extraction have been the main point of research and detection.To ensure the system security and stabilization and to force the power supply dependably and constantly, we need a high quality fault detection and diagnosis system on power supply line to identify the fault qualities of the traction feeding line correctly and quickly. This paper has been put forward in order to ensure the traction feeding line can supply steadily and realized the real-time fault detection and diagnosis on the feeding line state.Because permanent fault and transient fault can't be distinguished by the present protection mode, transmission line protection is compelled to deal with reclosure blindly, which bring unnecessary secondly damage to the power equipment and fault line. To solve the problem, a protection project based on fault traveling waves is brought forward in this paper, which can correctly identify the fault qualities and give a pulse signal to reclosure device, this ensure the reclosure's efficient action.There are some attributions: firstly, people can understand the fault traveling waves feature and transmission characteristic on power supply line in the paper. Based on the traveling waves feature, a new method has been put forward by author is that taking advantage of the wavelet transform to extract quality on the feeding line, the author use the ratio of wavelet energy in every frequency to signal total energy as quality parameter. At last, the fault feature vectors are classified correctly based on artificial neural nets, the conclusions are proved that protection project based on fault traveling waves can identify the traveling wave signals of different fault qualities, and achieve the object of fault pattern recognition.